Confident Splatting: Confidence-Based Compression of 3D Gaussian Splatting via Learnable Beta Distributions

TL;DR

This paper introduces a confidence-based lossy compression method for 3D Gaussian Splatting that reduces storage and computation by pruning low-confidence splats while maintaining visual quality, using learnable Beta distributions.

Contribution

It presents a novel, architecture-agnostic compression approach utilizing confidence scores modeled as Beta distributions, enabling effective pruning and quality assessment of 3D splats.

Findings

Achieves favorable compression-fidelity trade-offs

Uses average confidence as a new scene quality metric

Demonstrates effectiveness across various Gaussian Splatting architectures

Abstract

3D Gaussian Splatting enables high-quality real-time rendering but often produces millions of splats, resulting in excessive storage and computational overhead. We propose a novel lossy compression method based on learnable confidence scores modeled as Beta distributions. Each splat's confidence is optimized through reconstruction-aware losses, enabling pruning of low-confidence splats while preserving visual fidelity. The proposed approach is architecture-agnostic and can be applied to any Gaussian Splatting variant. In addition, the average confidence values serve as a new metric to assess the quality of the scene. Extensive experiments demonstrate favorable trade-offs between compression and fidelity compared to prior work. Our code and data are publicly available at https://github.com/amirhossein-razlighi/Confident-Splatting